Skip to content

shubham0204/NN_GeneticAlgo_Optimization_Kotlin

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Hyperparameter Optimization With Genetic Algorithms in Kotlin

image

Hyperparameter optimization is all about selecting the best hyperparameters like learning rate, number of layers, activation functions etc. for neural networks.

On the other hands, Genetic Algorithms ( GEs ) are Evolutionary Algorithms which work on the principle given by Charles Darwin, "Only the fittest individual survives". A population would contain a specific number of individuals ( NNs ), and the fittest one ( the one with the smallest loss ) will be evolved!

We have our project Coding Feed-forward Neural Networks in Kotlin ( GitHub , Medium ) and extending it further, we add GE based hyperparameter optimization to it, all in your favourite Kotlin.

Please Note!

Individuals who love linear algebra, matrics, partial derivatives and other weird stuff are warned. This repo has no math for you guys, but biology we learnt in our early classes